I microRNA (miRNA) tumorali sono nuovi biomarcatori promettenti per la diagnosi ed il monitoraggio dell'evoluzione tumorale. Possono essere individuati in una vasta gamma di campioni biologici e sono altamente stabili. È possibile valutare la loro espressione tramite le tecnologie Next-Generation Sequencing (NGS), che stanno portando ad un cambiamento nell'approccio ai complessi studi di genomica, inclusi quelli relativi all'RNA. Caratterizzate da un elevato throughput per sequenze relativamente corte (reads), sono il metodo più affidabile e preciso per raggruppare gli individui sulla base dei loro profili genetici. Nel presente studio, abbiamo analizzato i profili di espressione dei miRNA partendo da dati NGS in 2 tipi di campioni (feci e scrape cervicali) per identificare un pannello di miRNA informativo rispetto alla progressione tumorale. Proponiamo una pipeline di analisi dei dati integrata sotto il profilo sperimentale e computazionale. Dal punto di vista computazionale, abbiamo combinato algoritmi di mapping presenti nello stato attuale dell'arte con algoritmi sviluppati ad hoc. Abbiamo definito un workflow ottimizzato per l'analisi di reads ottenute da miRNA sequencing e la loro analisi di espressione quantitativa per individuare il set di miRNA differenzialmente espressi. Il nostro approccio è stato applicato all'analisi dei risultati preliminari di miRNA sequencing da 2 casi di studio, il cancro al colon retto (CRC) ed il cancro cervicale (CC), con un interesse mirato alla scoperta dei candidati biomarcatori. Abbiamo effettuato un primo studio pilota in campioni di feci provenienti da 12 soggetti (2 con CRC, 3 con adenomi, 2 aventi IBD (Inflammatory Bowel Disease) e 5 sani), per verificare il possibile uso di NGS per identificare i miRNA disregolati e dunque in grado di discriminare tra diversi stadi di CRC: lesioni precancerose (adenomi colorettali) o stati infiammatori (IBD). Nel secondo studio abbiamo analizzato 95 campioni di scrape cervicali (64 positivi al Virus del Papilloma umano (HPV) e 31 negativi) per identificare i miRNA in grado di discriminare tra le 3 lesioni di differente gravità della neoplasia intraepiteliale cervicale (CIN) (tra gli HPV positivi: 41 o senza lesione o con CIN1, 18 con CIN2 o CIN3 al reclutamento, 5 con CIN2 o CIN3 al follow-up). Nei campioni di feci, abbiamo individuato miRNA significativamente espressi in modo differenziale (DEmiRNAs) associati ad un fold change |> 3| ed un p-value corretto per Bonferroni <0,05: 3 DEmiRNAs in campioni con CRC (hsa-miR-6780b, hsa-miR-6857, hsa-miR-4327), 1 negli adenomi colorettali (hsa-miR-10b) e 1 in campioni con IBD (hsa-miR-300) rispetto ai campioni sani. Abbiamo valutato la presenza di miRNA appartenenti ad organismi non umani, che possono riflettere delle peculiarità individuali di stile di vita. Nello studio su scrape cervicali, abbiamo identificato 67 DEmiRNAs comuni ed i loro 5 target validati comuni (VEGFA, BCL2, BCL2L11, BMPR2, CCND1) confrontando I campioni HPV positivi con gli HPV negativi e gli HPV negativi con soggetti aventi lesioni CIN, utilizzando la stessa soglia sopra riportata. In questo studio abbiamo confrontato le coordinate dei miRNA precursori rispetto ai siti di integrazione di HPV. Dai nostri risultati preliminari, i miRNA possono differenziare pazienti con lesioni precancerose/cancerose da controlli sani, suggerendo una possibile direzione futura verso il loro uso per la diagnosi precoce/screening di tumori quali CRC e CC.
Tumour-related microRNAs (miRNAs) are promising novel biomarkers for diagnosis and monitoring of human cancers. miRNAs can be detected in a wide range of biological specimens, are stable and resistant to endogenous ribonuclease activity, extreme pH and temperature. Global expression of miRNAs can be assessed using Next-Generation Sequencing (NGS) technologies, which are rapidly changing the approach to complex genomic studies, including studies on whole RNA. These technologies, characterized by a massive throughput for relatively short sequences (reads), are currently the most reliable and accurate method for grouping individuals on the basis of their genetic profiles. In the present study, we were interested in analysing miRNA expression profiles by NGS in 2 different specimens (stool and cervical scrapes) for the identification of a broadly informative miRNA marker panel for cancer progression. We propose hereby a comprehensive approach from both the experimental and the computational points of view. In particular, for the latter, we have combined state-of-the-art mapping algorithms with ad-hoc developed algorithms. The optimized workflow for miRNA sequencing reads analysis and their quantitative expression analysis is used to obtain the set of differentially expressed miRNAs, the set of candidate biomarkers. Our approach was applied to the analysis of preliminary outcomes of miRNA sequencing from 2 studies, ColoRectal Cancer (CRC) and Cervical Cancer (CC) case studies, focused on biomarkers discovery. In the first, we have performed a pilot study in stool samples from 12 subjects (2 CRC, 3 Polyps, 2 Inflammatory, and 5 healthy) to test the possibility to use NGS in stool for the identification of dysregulated miRNAs able to discriminate different CRC stages, CRC precancerous lesions (adenomas) or inflammatory status (Inflammatory Bowel Disease, IBD). In the second study, we have analysed 95 cervical scrape samples (64 Human papilloma virus (HPV) positive and 31 HPV negative) to identify miRNAs able to discriminate 3 grades of Cervical Intraepithelial Neoplasia (CIN) precancerous lesions (between HPV positive samples: 41 with no lesion or CIN1, 18 with CIN2 or CIN3 lesions at recruitment, 5 with CIN2 or CIN3 lesions at follow up). In stool samples we have identified significantly differentially expressed miRNAs (DEmiRNAs) associated with a fold change |>3| and an adjusted p-value (with Bonferroni correction) < 0.05. We have found 3 DEmiRNAs in CRC samples (hsa-miR-6780b, hsa-miR-6857, hsa-miR-4327), 1 in colorectal adenomas (hsa-miR-10b) and 1 in IBD samples (hsa-miR-300) in comparison with healthy samples. We have also investigated the presence of miRNAs from non-human organisms, which may reflect individual lifestyle peculiarities. In the cervical scrapes study, we have identified 67 common DEmiRNAs and their 5 common validated targets (VEGFA, BCL2, BCL2L11, BMPR2, CCND1) comparing the HPV positive versus HPV negative and HPV negative versus CIN lesions using the same threshold reported above. In this study, we have compared the coordinates of precursor miRNAs with respect to HPV breakpoints integration sites. From our preliminary results, miRNAs investigated in easily collectable specimens may reliably differentiate patients with precancerous/cancerous lesions from healthy controls, suggesting a possible future direction towards their potential use for the early detection/screening of malignancies such as CRC and CC.
Analisi di sequenziamento di microRNA per l'identificazione di biomarcatori in due casi di studio, cancro al colon retto e cancro alla cervice uterina
GOUZOUNIS, STAVRULA
2014/2015
Abstract
Tumour-related microRNAs (miRNAs) are promising novel biomarkers for diagnosis and monitoring of human cancers. miRNAs can be detected in a wide range of biological specimens, are stable and resistant to endogenous ribonuclease activity, extreme pH and temperature. Global expression of miRNAs can be assessed using Next-Generation Sequencing (NGS) technologies, which are rapidly changing the approach to complex genomic studies, including studies on whole RNA. These technologies, characterized by a massive throughput for relatively short sequences (reads), are currently the most reliable and accurate method for grouping individuals on the basis of their genetic profiles. In the present study, we were interested in analysing miRNA expression profiles by NGS in 2 different specimens (stool and cervical scrapes) for the identification of a broadly informative miRNA marker panel for cancer progression. We propose hereby a comprehensive approach from both the experimental and the computational points of view. In particular, for the latter, we have combined state-of-the-art mapping algorithms with ad-hoc developed algorithms. The optimized workflow for miRNA sequencing reads analysis and their quantitative expression analysis is used to obtain the set of differentially expressed miRNAs, the set of candidate biomarkers. Our approach was applied to the analysis of preliminary outcomes of miRNA sequencing from 2 studies, ColoRectal Cancer (CRC) and Cervical Cancer (CC) case studies, focused on biomarkers discovery. In the first, we have performed a pilot study in stool samples from 12 subjects (2 CRC, 3 Polyps, 2 Inflammatory, and 5 healthy) to test the possibility to use NGS in stool for the identification of dysregulated miRNAs able to discriminate different CRC stages, CRC precancerous lesions (adenomas) or inflammatory status (Inflammatory Bowel Disease, IBD). In the second study, we have analysed 95 cervical scrape samples (64 Human papilloma virus (HPV) positive and 31 HPV negative) to identify miRNAs able to discriminate 3 grades of Cervical Intraepithelial Neoplasia (CIN) precancerous lesions (between HPV positive samples: 41 with no lesion or CIN1, 18 with CIN2 or CIN3 lesions at recruitment, 5 with CIN2 or CIN3 lesions at follow up). In stool samples we have identified significantly differentially expressed miRNAs (DEmiRNAs) associated with a fold change |>3| and an adjusted p-value (with Bonferroni correction) < 0.05. We have found 3 DEmiRNAs in CRC samples (hsa-miR-6780b, hsa-miR-6857, hsa-miR-4327), 1 in colorectal adenomas (hsa-miR-10b) and 1 in IBD samples (hsa-miR-300) in comparison with healthy samples. We have also investigated the presence of miRNAs from non-human organisms, which may reflect individual lifestyle peculiarities. In the cervical scrapes study, we have identified 67 common DEmiRNAs and their 5 common validated targets (VEGFA, BCL2, BCL2L11, BMPR2, CCND1) comparing the HPV positive versus HPV negative and HPV negative versus CIN lesions using the same threshold reported above. In this study, we have compared the coordinates of precursor miRNAs with respect to HPV breakpoints integration sites. From our preliminary results, miRNAs investigated in easily collectable specimens may reliably differentiate patients with precancerous/cancerous lesions from healthy controls, suggesting a possible future direction towards their potential use for the early detection/screening of malignancies such as CRC and CC.File | Dimensione | Formato | |
---|---|---|---|
736704_tesigouzounisstavrula.pdf
non disponibili
Tipologia:
Altro materiale allegato
Dimensione
2.84 MB
Formato
Adobe PDF
|
2.84 MB | Adobe PDF |
Se sei interessato/a a consultare l'elaborato, vai nella sezione Home in alto a destra, dove troverai le informazioni su come richiederlo. I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14240/11932